Deita Tech
A data and AI company building analytics and decision tools.
01
Selected publications
- 2026
- 2026
- 2026
Global Framework for Emulation of Nuclear Calculations
Journal Physical Review Letters 136, 082501
- 2025
Discovering Nuclear Models from Symbolic Machine Learning
Journal Communications Physics 8, 101
- 2023
A General Framework for Equivariant Neural Networks on Reductive Lie Groups
Proceedings NeurIPS 36 (2023)
02
Building
Qualia
A nonprofit working to make technology and science education accessible to more people.
03
Patents
Gated Attention Mechanism for Embedding Reduction of Tabular Data
Filed during my internship at Visa, alongside related published work on generative models for tabular data.
04
Skills
- Quantum many-body methods
- Ab initio approaches to nuclear structure (e.g. IM-SRG), shell-model calculations, and uncertainty quantification.
- Statistics & machine learning
- Bayesian inference, multi-fidelity emulation, Bayesian optimization, symbolic regression, equivariant ML, diffusion and generative models, interpretability and alignment.
- Computing
- Python, CUDA, PyTorch, JAX, C++, Julia, and large-scale scientific computing.
- Experiment
- Laser spectroscopy and data-acquisition system design for radioactive-beam experiments, control systems for precision measurements.
05
Talks & seminars
-
LNS Physics Colloquium
-
INT Workshop “Nuclear Hamiltonians for Advancing Nuclear Physics and Beyond”
-
Workshop on Experimental Design & Digital Twins — Instructor
-
DNP Meeting
-
Nuclear & Particle Physics Seminar
-
Enforcing Hard Constraints in Generative Models for Mixed-Type Tabular Data
-
BAyesian Neural Network: an Atomic and Nuclear Emulator (BANNANE)
-
What Is There Inside a BANNANE: ML Tools for Global Emulation
-
Ab Initio Nuclear Theory
-
AI4Physics
-
Synthetic Data vs. Traditional Methods: The Future of Market Research — Panelist